Accelerat ing t he world's research.

Risk Analysis and Bovine

Tuberculosis, a Re-emerging
Zoonosis
Alexandre Caron, E. Etter, Ferran Jori

Annals of the New York Academy of Sciences

Cite this paper Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers Download a PDF Pack of t he best relat ed papers 

Bovine t uberculosis: a double-edged issue at t he human/livest ock/wildlife int erface in Africa

Alexandre Caron, Michel de Garine-Wichat it sky

AN OVERVIEW ON BOVINE T UBERCULOSIS AND IT S ECONOMIC SIGNIFICANCE IN MEKELLE CIT Y, ET HIO…

Dr. Ahmed A L I Farah

Malama rev 13
Sydney malama
Risk Analysis and Bovine Tuberculosis, a
Re-emerging Zoonosis
ERIC ETTER,a PILAR DONADO,b FERRAN JORI,a
ALEXANDRE CARON,c FLAVIE GOUTARD,a AND FRANÇOIS ROGERa
a CIRAD,EMVT Department, Epidemiology and Ecology Unit,
34098 Montpellier, France
b University of California, Davis, California 95616, USA
c CIRAD,
EMVT Department, Integrated Wildlife Management Unit,
34398 Montpellier, France

ABSTRACT: The widespread of immunodeficiency with AIDS, the conse-

quence of poverty on sanitary protection and information at both indi-
vidual and state levels lead control of tuberculosis (TB) to be one of the
priorities of World Health Organization programs. The impact of bovine
tuberculosis (BTB) on humans is poorly documented. However, BTB re-
mains a major problem for livestock in developing countries particularly
in Africa and wildlife is responsible for the failure of TB eradication pro-
grams. In Africa, the consumption of raw milk and raw meat, and the
development of bushmeat consumption as a cheap source of proteins,
represent one of the principal routes for human contaminations with
BTB. The exploration of these different pathways using tools as partic-
ipatory epidemiology allows the risk analysis of the impact of BTB on
human health in Africa. This analysis represents a management support
and decision tool in the study and the control of zoonotic BTB.

KEYWORDS: bovine tuberculosis; risk analysis; participatory epidemiol-

ogy; zoonosis; bushmeat

INTRODUCTION

Human tuberculosis (TB) of animal origin (zoonotic TB) is an important

public health concern in developing countries. African nations pose a partic-
ular challenge in TB control, because of deficiencies in public health control
measures for cattle and animal products, coupled with a high prevalence and
incidence of HIV in the human population. Risk analysis is an important tool
in defining the public health danger posed by zoonotic TB in Africa. Using
risk analysis, two potential pathways are proposed as comprehensive models

Address for correspondence: François Roger, CIRAD, EMVT Department, Epidemiology and Ecol-
ogy Unit, 34098 Montpellier Cedex 5, France. Voice: +33 4 67 59 37 06; fax: +33 4 67 59 37 54.
e-mail: francois.roger@cirad.fr

Ann. N.Y. Acad. Sci. 1081: 61–73 (2006). 

C 2006 New York Academy of Sciences.

doi: 10.1196/annals.1373.006

61
62 ANNALS NEW YORK ACADEMY OF SCIENCES

when considering different routes of human infection: (a) from domestic and
wild animals; and (b) from animal products including milk and bushmeat.
Globally, wildlife reservoirs threaten the success of TB eradication pro-
grams (e.g., badgers in the United Kingdom;1 however, according to Donnelly
et al.,2 the presence of badgers seems actually to keep transmission of TB down
in cattle, possums in New Zealand, white-tailed deer in Michigan, USA, wild
boar and deer in Spain and France). In Africa, little is known about wildlife
reservoir, but in southern Africa (mainly South Africa), in the Kruger National
Park, 15 years of studies clearly indicate that the African buffalo (Syncerus
caffer) is the main reservoir of the pathogen agent, with other species like the
greater kudu (Tragelaphus strepsiceros) and common warthog (Phacochoerus
africanus) having a potential role in the continued prevalence of the disease.3
Depending on the specific ecosystem with respect to species abundance, diver-
sity, and anthropic factors, such as habitat impact and wildlife management,
wildlife species seem to be more or less susceptible to the Mycobacterium spp.
and therefore, the disease dynamics and the role of species have been varying
accordingly.
The purpose of this article is to examine the impact of bovine tuberculosis
(BTB) on human health in Africa and to propose an approach for its study and
its control using risk analysis.

ECOLOGY, EPIDEMIOLOGY, AND ZOONOTIC ASPECTS OF

BTB IN AFRICA

In Africa, BTB primarily affects cattle.4–9 However, infection in other farm

and domestic animals are sometimes reported. Mycobacterium bovis has a
broad host range as the principal cause of TB in free-living wildlife, cap-
tive wildlife, domestic livestock, and non-human primates.10 Wild ruminants
and carnivores, such as African buffalo, lion, cheetah, greater kudu, leop-
ard, warthog, and eland, can be infected.11–13 Scavengers (hyenas, genet) and
Chacma baboons in Kenya became infected through the ingestion of abattoir
wastes.14 Most of the time, BTB transmission is considered as passing from
livestock to wildlife. In wild ruminants, the disease has been documented
worldwide, and lesions and symptoms are very similar to those of domestic
ruminants.15
Depending on the susceptibility of a species to, and on the prevalence of BTB,
these animals could act as reservoirs or spill-over hosts to other species.13–16 In
reservoir or maintenance hosts, infection can persist through horizontal trans-
fer in the absence of any other source of M. bovis and may well be transmitted to
other susceptible hosts. Only a limited number of species act as maintenance or
reservoir hosts of TB, which include brush-tailed possum (Trichosurus vulpec-
ula), European badgers (Meles meles), bison (Bison bison), African buffalo
ETTER et al.: RISK ANALYSIS AND BOVINE TUBERCULOSIS 63

TABLE 1. Worldwide TB Figures (WHO 2002)

World population 6 219 000 000
TB-infected people 2 073 000 000
Estimated incidence Tuberculosis 8 797 000
Pulmonary TB (Smear-positive cases) 3 887 000
Deaths due to TB 1 800 000
TB cases due to HIV 9%
Deaths of TB patients due to HIV 12%

(Syncerus cafer), and white-tailed deer (Odocoileus virginianus). In contrast,

spill-over hosts become infected with M. bovis, but the infection only occurs
sporadically, or persists within these populations if a true maintenance host is
present in the ecosystem. In the case of high density populations, spill-over
hosts can act as reservoir hosts of M. bovis.17
In Africa, TB infection in humans is principally caused by Mycobacterium
tuberculosis. However, human TB of animal origin caused by M. bovis is be-
coming increasingly prevalent in developing countries18 due to the lack of
both control and diagnostic measures and pasteurization of milk.19,20 Con-
comitantly, TB is a major opportunistic infection in HIV-infected persons, and
the World Health Organization (WHO) estimated that 70% (6 million) of the
people co-infected with TB and HIV live in Sub-Saharan Africa.19 The preva-
lence, incidence, and deaths caused by TB, reported in 2002 by WHO, are
given in TABLE 1. Every year, there are 8–10 million new cases of TB reported,
and 2–3 million deaths attributed to TB,21 but the exact percentage of TB that
may be caused by M. bovis is not known. Global prevalence of human TB due
to M. bovis is estimated at 3.1% of all human TB cases, of which 2.1% are
pulmonary infections, and 9.4% extra pulmonary.19 However, the proportion
of M. Bovis in Africa and within the TB–HIV complex is unknown.
Information regarding human disease due to M. bovis is rare. In Africa, where
M. bovis is present in animal species, there is substantial lack of knowledge
about the distribution, epidemiological patterns, and transmission dynamics
of this important zoonosis. The fact that 50% of total African cattle happen
to be in countries without any control measures for BTB22 is a matter of con-
cern. Results from several studies conducted in different African countries
have clearly established the importance of BTB as a major public health prob-
lem. In Malawi, a survey of human sputum cultures from human TB patients
revealed that 42.8% of the culture-positive specimens were M. bovis (Ministry
of Health, 1985, cited by Wedlock et al.20 ). In Egypt, a study reported that 9
out of 20 randomly selected patients with TB peritonitis were infected with
M. bovis.23 Jiwa et al.,4 in Tanzania, suggested that the presence of BTB in
cattle necessitates further investigation into the role of animal-derived BTB in
human health. In Burkina Faso, the Ministry of Health reported 1,334 human
64 ANNALS NEW YORK ACADEMY OF SCIENCES

cases of zoonotic TB.24 Ameni et al.25 reported high occurrences of BTB in

cattle and cattle owners, respectively, in central Ethiopia, and pointed out the
need to assess and evaluate the impact of zoonotic TB on human populations
in order to design cost-effective control methods throughout the country. The
risk of contracting TB is 20–50 times greater in Africa than in Europe.18
The importance of M. bovis in human TB cases is not mentioned in any of
the national reports submitted to OIE (World Organization of Animal Health)
and WHO by African member states. Cases of M. bovis infection in humans are
underreported as a result of diagnostic limitations and non-adapting sampling
for searching extra pulmonary TB.19
TB infection may occur via contaminated materials (fomites), aerosolization
(human to human), and predation, or by vertical and/or horizontal transmission
in humans.15 Zoonotic TB caused by M. bovis could be acquired by humans
through the inhalation of cough sprayed from infected cattle, or from handling
or consumption of milk contaminated with the organism.20 BTB in humans
affects young individuals and causes cervical lymphadenopathy, intestinal le-
sions, chronic skin TB (Lupus vulgaris), and other non-pulmonary forms.19
If M. bovis is acquired by inhalation, humans typically develop pulmonary
TB.19
Another factor which increases the risk of acquiring zoonotic TB in Sub-
Saharan Africa is the active competition in all Sub-Saharan African countries
between large-scale commercial food enterprises and smaller, less-regulated
farmers who frequently ignore safety standards for hygiene and product quality.
These smaller farmers sell directly to final consumers and contribute to the
spread of TB.19 Furthermore, 90% of the total milk produced by these countries
and consumed by people is either fresh or soured, and not pasteurized (Walshe
et al., 1991, cited by Cosivi et al.19
Finally, we can underline that the recent development of wildlife activities,
such as game tourism, farming, and hunting to develop the peripheral zones
of protected areas contributes to the exposure of specific persons to Mycobac-
terium spp.

RESEARCH QUESTION, TOOLS, AND METHODOLOGIES

Efficient risk analysis could become an important tool in assessing and

overcoming the public health danger posed by zoonotic TB. Risk analysis is
a relatively new research tool that encompasses qualitative, deterministic, and
probabilistic health assessment. This tool requires interdisciplinary collabora-
tions to satisfy the demands for high quality characterization of risk. Risk could
be defined as the likelihood of occurrence of an adverse event and the sever-
ity of the consequences if the result does occur.26 MacDiarmid and Pharo27
defined risk analysis as a tool intended to provide decision-makers with an ob-
jective, repeatable, and documented appraisal of the risk posed by a particular
action.
ETTER et al.: RISK ANALYSIS AND BOVINE TUBERCULOSIS 65

Based on the OIE (World Organization for Animal Health) code, risk anal-
ysis constitutes four steps: (1) hazard identification; (2) risk assessment; (3)
risk management; and (4) risk communication.28 Risk assessment evaluates
the probability of entry, the establishment or spread of a disease under exist-
ing conditions, predetermined control measures, and the associated potential
biological and economic consequences of establishment of the disease. In the
case of zoonotic TB, it consists of the identification, estimation of statistical
probabilities, and evaluation of the consequences of all risks associated with
the transmission of BTB from animals to humans. Release assessment is the
probability of releasing viable M. bovis to bordering communities from (a)
cattle; (b) contaminated carcasses and meat; (c) contaminated milk; and (d)
bushmeat. Exposure assessment is the probability of being exposed to viable
M. bovis. Factors involved in this assessment are: (a) rural and peri-urban pop-
ulations; and (b) abattoir workers, farm workers, and people with a high degree
of contact with animals (e.g., game farm workers, rangers, wildlife vets). Con-
sumption of contaminated food and hunting habits are also exposure variables.
Consequence assessment is the probability of becoming infected, diseased, or
of dying. Consequences of human BTB are sickness, death, as well as the po-
tential for one group to be the release source for human-to-human transmission.
Risk analysis can be quantitative providing a numeric, or qualitative estimate,
when a descriptive approach is used.29 Both types of assessments are equally
valid if they are based on good quality data and concentrate on all the defined
stages of the infective process.29 Some of the inputs and variables of this model
are unknown, which requires that we incorporate uncertainty and variability
into the model. These variations are modeled using Monte-Carlo simulation
technique, a stochastic iterative approach.30 Input parameters should be de-
fined through the collection of existing data, expert opinions, and the use of
field surveys including participatory epidemiology (PE).
PE is an emerging field based on epidemiological technique using participa-
tory methods to collect epidemiological data by the widely accepted methods
of Rapid Rural Appraisal and Participatory Rural Appraisal. PE relies on ob-
servation, existing veterinary knowledge of traditional livestock owners, and
oral history from the local communities. PE is based on the principle of flex-
ibility with the use of iterative analysis and on the principle of triangulation
which is the cross-checking of information gained from several intentionally
different perspectives. A wide range of PE tools are available and can be cat-
egorized into three main groups, complemented with secondary sources and
direct observation in the field: informal interviews of key informants, visual-
ization methods, and ranking or scoring methods. Information derived from
all these sources is then combined and cross-checked to build a picture of
the issues under investigation. Use of conventional veterinary diagnostic tools
is an integral part of, and in some cases overlaps with PE methods. PE can
make use of quantitative information coming from previous formal epidemi-
ological surveys, and uses qualitative intelligence to fill the gaps between
66 ANNALS NEW YORK ACADEMY OF SCIENCES

available data.31 Examples of PE used in veterinary research include: basic

research on the epidemiology of endemic and epizootic diseases, used with
Orma communities in Kenya for the Bovine Trypanosomiasis,32 participatory
disease searching which is a widely applied method of assessing Rinderpest
risk, verifying eradication, and substantiating disease-free herd status,33 dis-
ease modeling done on the construction of a model of the contagious bovine
pleuropneumonia (CBPP) transmission in transhuman production systems.34
PE methods augment the capacities of conventional epidemiological methods
to provide reliable and fast epidemiological information on complex disease
concerns. For risk factor identification and assessment of the local epidemi-
ological situation of BTB, qualitative and semi-qualitative data are needed.
The use of PE methods could provide guidance on local attitudes and behavior
with regard to: animal husbandry practices, social and cultural habits (e.g., milk
processing, milk consumption, and hunting and consumption of bushmeat).35
Risk assessment for M. bovis human infection is exemplified using one
scenario pathway model: the milk pathway. This pathway determines an orderly
series of events which would ultimately lead to the acquisition of BTB by
humans in Uganda. The pathway looks at the probability that an individual is
infected with M. bovis at the source.
Proposals will be based on the milk and meat pathways in African countries.

FRAMEWORK PROPOSED FOR RISK ANALYSIS OF

ZOONOTIC BTB IN AFRICA MILK PATHWAY

In Uganda, as in most Sub-Saharan countries, human TB has a high preva-

lence. The Ugandan Ministry of Health reported in 1995 a mean annual in-
cidence rate of human TB of 1.34/1000 person-years, and the role of BTB is
unknown.9 The same authors, using intradermal TB-skin testing, reported a
74% prevalence of TB reactors among cattle herds, 6% of individual-animals,
and a within-herd range of 1–50% that could be 100% if suspicious reactors
were included.9 A regular high prevalence of BTB in cattle can be correlated
with the isolation of M. bovis in milk samples. Vekemans et al.36 reported
isolation of Mycobacteria in 26% of 60 retailed milk samples collected from
markets in Burkina Faso.
The high prevalence of BTB in Ugandan cattle could have public health
implications; however, there is no information about the risk of zoonotic in-
fections of BTB to humans. Because of this unknown transmission potential,
we are proposing a BTB risk analysis in order to make recommendations to
public health agencies for prevention and control of the disease. During their
6 years of study in the region, CIRAD (Centre de Coopération Internationale
en Recherche Agronomique pour le Développement) established a series of
comprehensive social and political networks. The public health authorities
are aware of CIRAD’s study and wish to understand the risk presented by
ETTER et al.: RISK ANALYSIS AND BOVINE TUBERCULOSIS 67

Milk production

Release Probability of
Assessment presence of viable
M. bovis in Milk
Milk distribution

Exposure Exposure of
Probability of
Assessment susceptible human human exposure
population

Establishing of
infection
Probability of
Consequence
infection, sickness
Assessment and death
Sickness/death (and spread)
(Spread to other
humans?)

FIGURE 1. Risk assessment for the milk pathway.

zoonotic TB. CIRAD has proposed a basis for a comprehensive quantitative

risk analysis (FIG. 1) with evidence of a potential risk, thorough knowledge
of the structure of the culture, networks within the population and the dairy
production industry, links to public health authorities and agencies, and orga-
nizing efficient field investigation capabilities. Risk hazard identification for
zoonotic TB is currently based on a cross-sectional study with TB patients of
the Mbarara Hospital in Uganda, in order to determine the risk factors for TB
caused by M. bovis. A laboratory study component includes bacterial isolation
of Mycoplasma spp. and characterization of M. bovis using molecular tech-
niques. Risk factors are being categorized using a questionnaire that accounts
for general information about the patients and their medical history, occupa-
tional information, household information, habits of milk consumption, habits
of meat consumption, and knowledge of TB transmission routes (including the
zoonotic aspects).

MEAT PATHWAYS WITH WILDLIFE AND DOMESTIC

ANIMALS APPROACH

Eating raw or undercooked meat is one of the ways of contracting BTB.

Because of local habits concomitant with a high prevalence of BTB in cat-
tle, some countries present a real zoonotic risk of BTB transmission. Ameni
et al. showed 99% of either raw meat or raw and cooked meat consump-
tion versus 1% of only cooked meat consumption in Central Ethiopia.25 The
same study showed herd and individual animal prevalences of BTB of 42.6%
and 7.9%, respectively. At the end, 24.5% of the interviewed households had
68 ANNALS NEW YORK ACADEMY OF SCIENCES

experienced at least one human TB case in the family. In Ethiopia, BTB preva-
lence is a major concern for veterinary services ranging from 3.4% in small-
holder production systems to 87% in intensive farms.37,38 Because of the lack
of sensibility of the routine post-mortem examination in the abattoirs and be-
cause of the high proportion of backyard slaughtering, this problem is in reality
a public health problem.
The bushmeat trade and meat consumption have been increasing at an alarm-
ing rate in Sub-Saharan Africa, particularly in the forested areas of West and
Central Africa.39 Its relative importance as a source of proteins as compared
to domestic animals and fish varies between 6% (southern Africa) and 55%
(Central Africa) of total protein consumption.40 This practice has demonstrated
that in areas where bushmeat is important, hunters and persons involved in the
bushmeat trade are exposed to the transmission of several wildlife diseases.41,42
Despite the possibilities of human TB infection through bushmeat, these pro-
tein products have rarely been considered as a source of TB. In order to assess
the risk of BTB infection from wildlife to humans, we need to investigate
the problem at different levels. Data on BTB in wildlife in most of West and
Central African countries are scarce or nonexistent. Proteins come from both
wildlife hunted in the neighboring forests, and from cattle herds arriving from
pastoral regions in several areas of Sub-Saharan Africa, such as the Central
African Republic and Cameroon. Because prevalence of TB is important in
transhuman herds,43 there is a risk of transmission from cattle to wildlife as
described in literature when wild and domesticated animals share the same ter-
ritory.11,17,19,44,45 A surveillance of BTB in bushmeat markets can be a method
to monitor the presence of M. bovis in wildlife species. An advantage for sam-
pling wildlife carcasses is that in bushmeat markets in West and Central Africa,
a variety of species are available. The choice of species to be examined will
depend on the likely prevalence of M. bovis within each species, their abun-
dance, and the ease of sampling different hosts.17 In some cases, spill-over
species such as carnivores and other species at the top of the food chain have
become useful sentinel species for the detection of BTB in other wildlife ani-
mals.46 The level of detection of BTB in wildlife species will also depend on
the kind of diagnosis used to test for the presence of the organism. In infected
white-tailed deer in Michigan, it has been demonstrated that acid fast staining
histopathology and PCR methods can underestimate the prevalence of BTB as
compared to culture methods.47 The chances of detecting M. bovis in a diseased
carcass are also variable depending on the anatomical site used for screening
the carcasses. The body parts harboring the most infectiveness will depend on
the pathogenesis of M. bovis in the species being tested. Most wild ruminants
present with gross lesions in the lymph nodes of the head,48 while mesenteric
lymph nodes lesions are more common in carnivores or scavengers.17
The risks of TB transmission from wildlife to humans have been insuffi-
ciently studied in Sub-Saharan areas where bushmeat consumption prevails.
Indeed, in those areas, contact with wildlife carcasses from a wide variety
ETTER et al.: RISK ANALYSIS AND BOVINE TUBERCULOSIS 69

of species is probably more widespread and common than in southern Africa.

There, the most likely source of infection might be the cutaneous route, because
slaughtering is performed in open areas. On the other hand, overcooking meat
is a common measure used to prevent gastro-intestinal infections in African
countries, and venison is seldom consumed insufficiently cooked. However,
socioeconomic conditions, such as a lack of hygiene, informal trade, lack
of veterinary inspection and food safety services, lack of disease awareness,
and not using gloves, can be considered as risk factors for bushmeat traders,
butchers, hunters, and middlemen to become infected via the cutaneous route,
particularly when they cut themselves while dressing the carcass, or when they
dress an unprotected, open wound or abrasion on their arms or forearms.49
Consequences for humans infected with M. bovis through bushmeat path-
ways are probably very similar to those assessed for milk consumers. The
difference might be that the most exposed persons are adults, but with milk
consumers, children are also exposed. Women are usually involved in the bush-
meat trade and marketing, while men are more involved in hunting activities.
Dressing carcasses is more commonly done in the field by men, who deal
with heavy species (e.g., antelopes, bushpigs), while small species, such as
primates, rodents, and reptiles are more likely to be dressed by women in the
markets.
In addition, some molecular characterization techniques—for example,
spacer oligonucleotide typing,50 Single-step DNA-based assay51 RFLP52 —if
tested specifically for M. bovis in multi-species model could permit, if suc-
cessful, to introduce an idea of the historical contingency of disease strains
in wildlife and domestic populations and at their interface. From these links,
hypothetical pathways can be proposed and tested.
Last, the understanding of BTB dynamics in and between wildlife and do-
mestic reservoirs is an important issue to be looked at. Pathogens’ dynamics
between the wildlife and domestic reservoirs in some areas will explain the
incidence of the disease in cattle and therefore, will have an impact on the risk
assessment in humans. Later in the analysis, risk management can take into ac-
count wildlife reservoir management options, inducing a change in prevalence
in the domestic reservoir and changing the risk exposure of human population
at a higher level.

CONCLUSION

Despite the lack of study coverage related to the impact of BTB in human
TB cases, one can hypothesize that immunodepressed HIV patients would
be more susceptible to BTB contamination when a specific threshold of risk
factors is encountered. Contamination from human to human of M. bovis is
usually considered as a rare event. But in the context of high HIV prevalence
and the presence of many risk factors, the epidemiology of the disease in the
70 ANNALS NEW YORK ACADEMY OF SCIENCES

human host could be changed and human-to-human transmission could be

more common.43
The increase of population in Africa is leading to the use of more and more
marginal lands by pastoral societies and agricultural “colons.” These marginal
lands correspond to areas where the remaining population of wildlife has been
pushed by land-use pressure. Most of the national parks in Africa are based
in these areas. Therefore, one can observe an increased contact between live-
stock production systems and wildlife ecosystems, leading to the potential
for pathogen transmission in both ways. Another consequence linked to the
political agenda trying to integrate development and conservation issues in
the areas surrounding the national parks is the development sustainable activ-
ities, including game hunting and sustainable natural resources harvest. These
activities also constitute new risk factors for the transmission of pathogens.
Thus, contamination from buffalo and other wildlife species to humans has
to be taken into account in terms of bushmeat consumption, as also includ-
ing game hunting and traditional hunting. Other potential pathways should be
studied and included as direct transmission from livestock. Modeling of the
interface between wild and domestic animals will also help in building up
comprehensive risk assessment models.

REFERENCES

1. GORMLEY, E. & J.D. COLLINS. 2000. The development of wildlife control strategies
for eradication of tuberculosis in cattle in Ireland. Tuber. Lung Dis. 80: 229–236.
2. DONNELLY, C.A. et al. 2003. Impact of localized badger culling on tuberculosis
incidence in British cattle. Nature 426: 834–837.
3. BENGIS, R.G. et al. 2004. The role of wildlife in emerging and re-emerging
zoonoses. Rev. Sci. Tech. 23: 497–511.
4. JIWA, S.F. et al. 1997. Bovine tuberculosis in the Lake Victoria zone of Tanzania
and its possible consequences for human health in the HIV/AIDS era. Vet. Res.
Commun. 21: 533–539.
5. OMER, M.K. et al. 2001. A cross-sectional study of bovine tuberculosis in dairy
farms in Asmara, Eritrea. Trop. Anim. Health Prod. 33: 295–303.
6. BONSU, O.A., E. LAING & B.D. AKANMORI. 2000. Prevalence of tuberculosis in
cattle in the Dangme-West district of Ghana, public health implications. Acta
Trop. 76: 9–14.
7. KAZWALA, R.R. et al. 1998. Isolation of Mycobacterium species from raw milk of
pastoral cattle of the Southern Highlands of Tanzania. Trop. Anim. Health Prod.
30: 233–239.
8. MARTRENCHAR, A. et al. 1993. Problems associated with tuberculosis and brucel-
losis skin-test methods in northern Cameroon. Prev. Vet. Med. 15: 221–229.
9. BERNARD, F. et al. 2005. Tuberculosis and brucellosis prevalence survey on dairy
cattle in Mbarara milk basin (Uganda). Prev. Vet. Med. 67: 267–281.
10. O’REILLY, L.M. & C.J. DABORN. 1995. The epidemiology of Mycobacterium bovis
infections in animals and man: a review. Tuber. Lung Dis. 76(Suppl 1): 1–46.
ETTER et al.: RISK ANALYSIS AND BOVINE TUBERCULOSIS 71

11. WOODFORD, M.H. 1982. Tuberculosis in wildlife in the Ruwenzori National Park,
Uganda (Part II). Trop. Anim. Health Prod. 14: 155–160.
12. KEET, D.F. et al. 1996. Tuberculosis in buffaloes (Syncerus caffer) in the Kruger
National Park: spread of the disease to other species. Onderstepoort J. Vet. Res.
63: 239–244.
13. MICHEL, A.L. 2002. Implications of tuberculosis in African wildlife and livestock.
Ann. N. Y. Acad. Sci. 969: 251–255.
14. TARARA, R. et al. 1985. Tuberculosis in wild olive baboons, Papio cynocephalus
anubis (Lesson), in Kenya. J. Wildl. Dis. 21: 137–140.
15. BIET, F. et al. 2005. Zoonotic aspects of Mycobacterium bovis and Mycobacterium
avium-intracellulare complex (MAC). Vet. Res. 36: 411–436.
16. DELAHAY, R.J., C.L. CHEESEMAN & R.S. CLIFTON-HADLEY. 2001. Wildlife disease
reservoirs: the epidemiology of Mycobacterium bovis infection in the European
badger (Meles meles) and other British mammals. Tuberculosis (Edinb.) 81:
43–49.
17. DE LISLE, G.W. et al. 2002. Tuberculosis in free-ranging wildlife: detection, diag-
nosis and management. Rev. Sci. Tech. 21: 317–334.
18. BEDARD, B.G., S.W. MARTIN & D. CHINOMBO. 1993. A prevalence study of bovine
tuberculosis and brucellosis in Malawi. Prev. Vet. Med. 16: 193–205.
19. COSIVI, O. et al. 1998. Zoonotic tuberculosis due to Mycobacterium bovis in de-
veloping countries. Emerg. Infect. Dis. 4: 59–70.
20. WEDLOCK, D.N. et al. 2002. Control of Mycobacterium bovis infections and the
risk to human populations. Microbes Infect. 4: 471–480.
21. DAVIES, P.D. & J.M. GRANGE. 2001. Factors affecting susceptibility and resistance
to tuberculosis. Thorax 56( Suppl 2): ii23–ii29.
22. WHO. 1994. TB: a Global Emergency, WHO Report on the TB Epidemic/WHO
Global Tuberculosis Programme. World Health Organization. Geneva, Switzer-
land.
23. NAFEH, M.A. et al. 1992. Tuberculous peritonitis in Egypt: the value of laparoscopy
in diagnosis. Am. J. Trop. Med. Hyg. 47: 470–477.
24. COULIBALY, N.D. & K.R. YAMEOGO. 2000. Prevalence and control of zoonotic dis-
eases: collaboration between public health workers and veterinarians in Burkina
Faso. Acta Trop. 76: 53–57.
25. AMENI, G., K. AMENU & M. TIBBO. 2003. Bovine tuberculosis: prevalence and
risk factor assessment in cattle and cattle owners in Wuchale-Jida District,
Central Ethiopia. Int. J. Appl. Res. Vet. Med. http://www.jarvm.com/articles/
Vol1Iss1/AMENIJVM.htm
26. NORTH, D.W. 1995. Limitations, definitions, principles and methods of risk anal-
ysis. Rev. Sci. Tech. 14: 913–923.
27. MACDIARMID, S.C. & H.J. PHARO. 2003. Risk analysis: assessment, management
and communication. Rev. Sci. Tech. 22: 397–408.
28. OIE. 2002. International Animal Health Code: Mammals, Birds and Bees. OIE.
Paris.
29. ZEPEDA, C. 2004. Risk Communication. OIE Conference. Panama. http://www.
oie.int/downld/Panama riskcom nov04.pdf
30. VOSE, D.J. 2000. Risk Analysis—A Quantitative Guide. John Willey & Sons, Ltd.
New York.
31. CATLEY, A. & B. ADMASSU. 2003. Using participatory epidemiology to assess
the impact of livestock diseases. FAO-OIE-AU/IBAR-IAEA Consultative Group
Meeting on Contagious Bovine Pleuropneumonia in Africa.
72 ANNALS NEW YORK ACADEMY OF SCIENCES

32. CATLEY, A. & P. IRUNGU. 2000. Participatory research on bovine trypanosomo-

sis in Tana river district, Kenya: preliminary findings and identification of
best–best interventions. PAVE Project and Kenya Trypanosomiasis Research
Institute, Nairobi, Kenya. http://www.participatoryepidemiology.info/Tana%
zoRiver%zoresearch.pdf
33. AU/IBAR. 2001. Rinderpest eradication strategy in the West and East Nile ecosys-
tems. Community-based Animal Health and Participatory. Epidemiology Unit,
African Union/Interafrican Bureau for Animal Resources. Nairobi.
34. AU/IBAR. 2002. Contagious bovine pleuropneumonia in pastoralist areas of East
Africa: disease dynamics and control options. African Union/Interafrican Bureau
for Animal Resources. Nairobi.
35. ROBIN, R.A., A. CATLEY & D. HIRD. 2004. Significance of participatory epi-
demiology in veterinary public health community-based. Expert Consultation
on Community-Based Veterinary Public Health Systems FAO 2: 27–37.
36. VEKEMANS, M. et al. 1999. Potential source of human exposure to Mycobacterium
bovis in Burkina Faso, in the context of the HIV epidemic. Clin. Microbiol.
Infect. 5: 617–621.
37. REDI, N. 2003. Prevalence of bovine tuberculosis and zoonotic implication in
Asela Twon, South East, Ethiopia. Doctorate of Veterinary Medecine Thesis,
Addis Ababa University, Debre-Zeit.
38. AMENI, G., P. BONNET & M. TIBBO. 2003. A cross-sectional study of bovine
tuberculosis in selected dairy farms in Ethiopia. Int. J. Appl. Res. Vet. Med.
http://www.jarvm.com/articles/Vol.Iss4/Tibbo.htm
39. ROBINSON, J.G. & E.L. BENNETT. 2000. Hunting for Sustainability in Tropical
Forests. Columbia University. New York.
40. CHARDONNET, P. et al. 2002. The value of wildlife. Rev. Sci. Tech. 21: 15–51.
41. KRUSE, H., A.M. KIRKEMO & K. HANDELAND. 2004. Wildlife as source of zoonotic
infections. Emerg. Infect. Dis. 10: 2067–2072.
42. LEROY, E.M. et al. 2004. Multiple Ebola virus transmission events and rapid decline
of Central African wildlife. Science 303: 387–390.
43. AYELE, W.Y. et al. 2004. Bovine tuberculosis: an old disease but a new threat to
Africa. Int. J. Tuberc. Lung Dis. 8: 924–937.
44. GALLAGHER, J. & R.S. CLIFTON-HADLEY. 2000. Tuberculosis in badgers: a review
of the disease and its significance for other animals. Res. Vet. Sci. 69: 203–
217.
45. GORTAZAR, C. et al. 2005. Molecular characterization of Mycobacterium tubercu-
losis complex isolates from wild ungulates in south-central Spain. Vet. Res. 36:
43–52.
46. CARON, A., P.C. CROSS & J.T. DU TOIT. 2003. Ecological implication of bovine
tuberculosis in African buffalo herds. Ecol. Appl. 13: 1338–1345.
47. O’BRIEN, D.J. et al. 2004. Estimating the true prevalence of Mycobacterium bovis
in hunter-harvested white-tailed deer in Michigan. J. Wildl. Dis. 40: 42–52.
48. SCHMITT, S.M. et al. 1997. Bovine tuberculosis in free-ranging white-tailed deer
from Michigan. J. Wildl. Dis. 33: 749–758.
49. WILKINS, M.J. et al. 2003. Mycobacterium bovis (bovine TB) exposure as a recre-
ational risk for hunters: results of a Michigan Hunter Survey, 2001. Int. J. Tuberc.
Lung Dis. 7: 1001–1009.
50. ARANAZ, A. et al. 1996. Spacer oligonucleotide typing of Mycobacterium bo-
vis strains from cattle and other animals: a tool for studying epidemiology of
tuberculosis. J. Clin. Microbiol. 34: 2734–2740.
ETTER et al.: RISK ANALYSIS AND BOVINE TUBERCULOSIS 73

51. COETSIER, C. et al. 2000. Duplex PCR for differential identification of Mycobac-
terium bovis, M. avium, and M. avium subsp. paratuberculosis in formalin-fixed
paraffin-embedded tissues from cattle. J. Clin. Microbiol. 38: 3048–3054.
52. BLAZQUEZ, J. et al. 1997. Genetic characterization of multidrug-resistant Mycobac-
terium bovis strains from a hospital outbreak involving human immunodeficiency
virus-positive patients. J. Clin. Microbiol. 35: 1390–1393.